•   When: Monday, November 28, 2016 from 10:00 AM to 12:00 PM
•   Speakers: Gene Shuman
•   Location: ENGR 4801
•   Export to iCal

People want to live independently, but too often disabilities or advanced age robs them of the ability to do the necessary activities of daily living (ADLs).  Finding relationships between electromyograms measured in the arm and movements of the hand and wrist needed to perform ADLs can help address performance deficits and be exploited in designing myoelectrical control systems for prosthetics and computer interfaces.

This dissertation presents the results of applying several machine learning techniques to discover the electromyogram patterns present when performing typical fine motor functional activities used to accomplish ADLs.  The primary data in this research is from electromyogram and accelerometer signals collected from the arms and hands of several subjects while they performed typical ADLs involving grips or movements of the hand and wrist. Four approaches were developed and tested.  One involved classification of 100 ms individual signal instances.  Two used nearest neighbor classification in two specific situations: creation of an affinity matrix to model learning instances and classify based on multiple adjacent signal values, and using Dynamic Time Warping (DTW) as a distance measure to classify entire activity segments.  These two techniques used a symbolic representation called SAX to approximate signal streams.  A fourth approach was developed to test continuous movement segments to more precisely labeled data.  It classified a signal instance by applying a 'belief' calculation that uses that instance's signal reading, the predicted class of the previous reading, and estimated transition probabilities.  Accelerometer data were systematically used to aid in labeling the data since it clearly indicates the start and stop of dynamic movements.